We conducted a randomized trial to evaluate primarily the cardioprotective effect of dexrazoxane (DEX) in patients with advanced breast cancer and soft tissue sarcomas (STS) treated with high-dose epirubicin (EPI). We wished also to determine the value of radioimmunoscintigraphy (RIS) in the assessment of anthracycline cardiotoxicity. Patients with breast cancer (n = 95) or STS (n = 34) received EPI 160 mg/m2 by intravenous (I.V.) bolus every 3 weeks with or without DEX 1,000 mg/m2 I.V. Cardiac monitoring included multigated radionuclide (MUGA) scans with determination of resting left ventricular ejection fraction (LVEF), and RIS with indium 111 antimyosin monoclonal antibodies. In either disease, antitumor response rates, time to progression, and survival did not significantly differ between the two arms. There was little difference in noncardiac toxicity for the two treatment groups. All methods of cardiac evaluation clearly documented the cardioprotective effect of DEX. Four patients developed congestive heart failure (CHF), all in the EPI arm. The decrease in LVEF from baseline was significantly greater in the control group. An abnormal antimyosin uptake was observed early in both arms and progressively increased during treatment. However, this increase was significantly higher in the EPI group (P = .004). DEX significantly protects against the development of cardiotoxicity when high single doses of EPI are used. Apparently, there was no evidence of an adverse impact of DEX on antitumor activity. Although RIS is a sensitive technique in detecting anthracycline cardiac damage, its specificity is low and it cannot be considered a primary test for guiding anthracycline treatment.